Chandler · usage
← Back to sessions

Session

User
yordan@dext.com · Yordan Todorov Ivanov
Started
2026-09-10 10:31
Last activity
2026-09-10 10:31
Questions
1
Grouping
time_gap — inferred — consecutive questions from this user with no long gap
session_id
ac099fa3-f528-4968-8a64-d1d9640ce920

Transcript

  1. 2026-09-10 10:31 ask success $0.0275 14.1 s full detail →
    Which Omni field tracks document extraction usage that counts toward an account's "prepare" plan limit? I'm seeing options like Volume of Receipts, Volume of Items Submitted, and Volume of Items Extracted - which one is the correct field for usage against the prepare limit?
    The documented field for extraction usage is **Volume of Items Extracted**: `omni_dbt__eds_prepare_account_user_event_activity.item_extracted_volume_total`. It sums items whose fields were set in Dext’s Extraction Sequence. Submission and receipt-volume measures track submissions, not extraction. However, the documentation does not explicitly confirm that this measure is enforced against the Prepare plan limit. - **Field:** `omni_dbt__eds_prepare_account_user_event_activity.item_extracted_volume_total` - **Aggregation:** Sum of `item_extracted_volume` - **Logic:** Use extraction volume rather than receipt or overall submission volume. - **Documentation gap:** The link between this field and the contractual Prepare plan limit is not provided; confirm with the Analytics team. **Warning:** This answer was generated by AI and should be double-checked.

Cost is the LLM completion spend LiteLLM priced for each call, summed per request. Chandler is running on the Codex CLI, which draws ChatGPT plan quota rather than per-token API billing, so this figure is not money paid — it is what the same traffic would have cost on the API, priced from the token counts Codex reports. Read it as the size of the bill avoided. Two known limits: embedding spend is not recorded, so retrieval and matching cost is missing, and because the figure is one sum per request it cannot be split by model within a request — a request's classifier and agent calls can use different models while llm_model holds only one name. Only authenticated calls are logged, USAGE_LOG_ENABLED can switch logging off, and log writes are fail-soft — this is not a complete record of traffic. Questions are grouped into sessions: a session is exact when the caller echoed its id back to us and otherwise inferred from a 30-minute gap in that user’s activity, so a grouping is only as good as the source shown on the session itself. A call with no attributable user gets no session at all; those questions are listed separately rather than dropped.